guile-user
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: The TTN dilemma (was Re: guile-www)


From: Chris Hall
Subject: Re: The TTN dilemma (was Re: guile-www)
Date: Sun, 06 Jun 2004 02:31:01 -1000
User-agent: Gnus/5.1006 (Gnus v5.10.6) Emacs/21.2 (gnu/linux)

Thamer Al-Harbash <address@hidden> writes:
>
>> The little bit I can glean as to the 'why' of the fork from TTN's 'glug'
>> web page is that 'the fork' was *because* of a backward compatability
>> issue?
>
> I'm guessing he didn't like the SCM changes. guile went through
> some major changes between 1.4 and 1.6

Seems a safe bet - the 'glug' version is at 1.4.1.  I've asked ttn for
clarification - it seems reasonable to me that people know why and how
they are different with out having to examine the code for the
interpreter itself, or some such similar effort.

>
>> Does anyone else find it ironic (as well as a PITA) that this seems to
>> have led to many more, but undocumented, incompatabilities?
>
> Absolutely. I actually considered moving to other
> interpreters. It's just a bad thing for the community. Seriously,
> I'd like to contribute _without_ the politics. The second people
> start acting deceptive is the second I lose interest. guile is
> good without all the bindings. You can usually roll these as you
> go :)
>

I like these comments!  Life is too short and there is too much code
waiting to be written for us to be quibbling about our tools!

Often, one of the greatest attractions of a particular piece of free
software *is* the community using that tool, and my sense is that this
is important to Guile users, and that they are in fact troubled by these
developments.  And conflicted - ttn does some very useful stuff, people
generally agree on that.

Honest, open differences are fine, and normal, and to be expected - but
what is the point if no one knows the substance of, and reasons behind,
the differences?

>
> version which worked fine with the official guile. The only
> problem it had, which I remember, is grokking the inet address
> values that pgsql supports. It was a simple fix when I fixed it.

I've gotten Guile-PG 0.16 working with 1.6.4, with a little help - had
to fiddle with -rpath link setting in the Makefile, nothing major.

>
> Nail. Hammer. Head.

;-) Naaaw - I *love* spending many wasted hours on things that easily
  could have been avoided.

>
> The only way to solve this is to have an effort to take all his
> code and make it compatible. It's a lot of effort, and work which
> is better spent _doing_ things with guile. Patching someone elses
> bindings to be compatible is one massive waste of time. You could
> probably re-write most of guile-pg in one night.

Right, right, right, and right.  But sheesh!

Well, maybe not the _only_ way - see below.

>
> Here's a problem though, and I'd love it if someone could tell me
> how to solve it. If I wrote a guile-pg, guile-sdl, and a bunch of
> other bindings, how do I take the namespace back from TTN?
> Writing bindings isn't very hard and I have some spare time this
> month.

guile-tng-pg?  guile-pg-tng?  the-one-true-guile-pg?
guile-gnu-pg?  gnu-guile-pg?  guile-pg-gnu?  guile-pg-rms?

ttn offered to consider a patch, if I cared to submit one, that would
allow guile-pg 0.19 to build under guile 1.6.  Which got me thinking
(though it is sometimes suggested I best avoid this activity).

So, I've also asked ttn if the incompatibilities are/could be generally
isolated somehow so that the major compatibility issues for guile
extensions could somehow be addressed via build.  You might want to look
at my post in response to his response to my question. :-)

Aloha,
+Chris

-- 
The law is not so much carved in stone as it is written in water,
flowing in and out with the tide.
-- Jeff Melvoin, Northern Exposure, Crime and Punishment, 1992

Attachment: pgpLeqHrXddKu.pgp
Description: PGP signature


reply via email to

[Prev in Thread] Current Thread [Next in Thread]